Liver cirrhosis in co-infected: prognosis poor, but improved by HAART

In a retrospective cohort of patients at San Matteo hospital in Pavia, Italy, cirrhosis of the liver was associated with significant morbidity and mortality in people coinfected with HIV and hepatitis B or C within five years of diagnosis, but the use of antiretrovirals led to significantly higher survival rates. The findings were published in the November 1^st, 2007 issue of the Journal of Acquired Immune Deficiency Syndromes.

Progressive chronic liver disease is common in people co-infected with HIV and hepatitis B virus (HBV) and/or hepatitis C virus (HCV). In the initial stage, compensated cirrhosis, liver tissue is scarred but metabolic function still partially compensates for the damage.

Compensated cirrhosis may progress to decompensated cirrhosis, marked by the development of jaundice, haemorrhage, ascites (build-up of abdominal fluid) or hepatitis-related encephalopathy. Coinfection is associated with a poorer prognosis, but relatively few studies to date have described the progression of end-stage liver disease in the coinfected population.

This study team retrospectively identified a cohort of coinfected adults with initially compensated liver cirrhosis, and tracked the time to decompensated cirrhosis and death. Two control groups, one without cirrhosis, and one without viral hepatitis infection, were also analysed. For the six-year period between January 1999 and December 2004, the team followed 140 adults with HIV infection only, 183 with HCV coinfection but no liver cirrhosis, and 69 with HBV- or HCV-related compensated cirrhosis. Of the 69 cirrhotic patients, 59 had HCV, seven had HBV, and three were coinfected with all three viruses. None of the participants were receiving any anti-HCV therapy.

Each participant was followed for at least six months, with assessments every three months during their regular clinical care, and an abdominal ultrasound every six months. All coinfected patients were biopsied at the beginning of follow-up to grade the extent of liver damage. Median age was between 35 and 37 years, participants were mostly male. CD4 cell counts were below 350 cells/mm³ in 88% of the cirrhotic patients, 53% of the non-cirrhotic, and 51% of non-co-infected controls. Cirrhotic patients tended not to be on antiretroviral therapy: 58% were receiving anti-HIV treatment, compared to 74% to 75% in the two control groups.

As would be expected, prognosis was much poorer for the patients with advanced liver disease. The five-year survival rate was 71.5% for cirrhotic patients and 97.1% for the patients without cirrhosis. During the entire follow-up period, 22 (31.9%) of the cirrhotic patients died, all of complications from liver disease. Eight (4.4%) and five (3.6%) of the coinfected and monoinfected control patients died, respectively; none of these deaths were liver-related.

The patients with liver cirrhosis at the start of follow-up had a mortality rate of 71.3 per 1000 person-years (95% confidence interval [CI], 47 to 108). The mortality rate was 8 (95% CI, 4 to 16) for co-infected patients without cirrhosis, and 6.5 (95% CI, 2.7 to 15.5) in patients with HIV infection only.

The most frequent complications were ascites, seen in 39.1% of the cirrhotic patients, jaundice in 39.1%, and hepatic encephalopathy in 21.7%; other studies have also noted ascites as a frequent complication and marker of progression in coinfected people. However, hepatocellular carcinoma (HCC) occurred in only 13% of this cohort; not as commonly as in other studies. All three of the triply-infected patients developed HCC during follow-up.

After the first decompensation-defining event, the survival rate was 48% at one year and 18.1% at three years. Antiretroviral treatment after the first event greatly improved survival. At one year, survival rates were 61.1% for those who received anti-HIV therapy and 26.7% for those who did not (p < 0.0001). The authors note that results may be biased by the retrospective study design, relatively small sample, and because people with better liver function may have been better able to continue antiretrovirals. Neither CD4 count nor HIV RNA levels were significantly associated with decompensation or death.

Although the study did not follow an HCV-monoinfected, HIV-negative group, the authors cited data from previous studies of this population for comparison. One long-term retrospective study of HCV-monoinfected individuals with compensated cirrhosis found a 7% annual rate of progression to decompensated cirrhosis, with 91% survival at five years and 79% at ten years (Fattovich 1997). In contrast, the current study found a five-year survival rate of only 71.5% in co-infected individuals with compensated cirrhosis.

The researchers conclude that "significant morbidity and mortality occurs during the first 5 years after the diagnosis of compensated cirrhosis in HIV-coinfected patients", but that "the maintenance of [antiretroviral] therapy after first decompensation was associated with improved survival… suggest[ing] that strategies for early intervention are needed to prevent disease progression in HIV-infected patients with concurrent HBV and/or HCV infection."